Advances in Artificial Intelligence: 25th Canadian Conference on Artificial Intelligence, Canadian AI 2012, Toronto, ON, Canada, May 28-30, 2012. Proceedings PDF
Preview Advances in Artificial Intelligence: 25th Canadian Conference on Artificial Intelligence, Canadian AI 2012, Toronto, ON, Canada, May 28-30, 2012. Proceedings
Lecture Notes in Artificial Intelligence 7310 Subseries of Lecture Notes in Computer Science LNAISeriesEditors RandyGoebel UniversityofAlberta,Edmonton,Canada YuzuruTanaka HokkaidoUniversity,Sapporo,Japan WolfgangWahlster DFKIandSaarlandUniversity,Saarbrücken,Germany LNAIFoundingSeriesEditor JoergSiekmann DFKIandSaarlandUniversity,Saarbrücken,Germany Leila Kosseim Diana Inkpen (Eds.) Advances in Artificial Intelligence 25th Canadian Conference on Artificial Intelligence, Canadian AI 2012 Toronto, ON, Canada, May 28-30, 2012 Proceedings 1 3 SeriesEditors RandyGoebel,UniversityofAlberta,Edmonton,Canada JörgSiekmann,UniversityofSaarland,Saarbrücken,Germany WolfgangWahlster,DFKIandUniversityofSaarland,Saarbrücken,Germany VolumeEditors LeilaKosseim ConcordiaUniversity DepartmentofComputerScienceandSoftwareEngineering FacultyofEngineeringandComputerScience Montreal,H3G1M8,QC,Canada E-mail:[email protected] DianaInkpen UniversityofOttawa SchoolofElectricalEngineeringandComputerScience FacultyofEngineering Ottawa,K1N6N5,ON,Canada E-mail:[email protected] ISSN0302-9743 e-ISSN1611-3349 ISBN978-3-642-30352-4 e-ISBN978-3-642-30353-1 DOI10.1007/978-3-642-30353-1 SpringerHeidelbergDordrechtLondonNewYork LibraryofCongressControlNumber:2012937365 CRSubjectClassification(1998):I.3,H.3,I.2.7,H.4,F.1,H.5.2,I.4-5 LNCSSublibrary:SL7–ArtificialIntelligence ©Springer-VerlagBerlinHeidelberg2012 Thisworkissubjecttocopyright.Allrightsarereserved,whetherthewholeorpartofthematerialis concerned,specificallytherightsoftranslation,reprinting,re-useofillustrations,recitation,broadcasting, reproductiononmicrofilmsorinanyotherway,andstorageindatabanks.Duplicationofthispublication orpartsthereofispermittedonlyundertheprovisionsoftheGermanCopyrightLawofSeptember9,1965, initscurrentversion,andpermissionforusemustalwaysbeobtainedfromSpringer.Violationsareliable toprosecutionundertheGermanCopyrightLaw. Theuseofgeneraldescriptivenames,registerednames,trademarks,etc.inthispublicationdoesnotimply, evenintheabsenceofaspecificstatement,thatsuchnamesareexemptfromtherelevantprotectivelaws andregulationsandthereforefreeforgeneraluse. Typesetting:Camera-readybyauthor,dataconversionbyScientificPublishingServices,Chennai,India Printedonacid-freepaper SpringerispartofSpringerScience+BusinessMedia(www.springer.com) Preface This volume contains the papers presented at the 25th Canadian Conference on Artificial Intelligence (AI 2012). The conference was held at York University in Toronto, Ontario, during May 28–30, 2012, and was collocated with the 38th Graphics Interface Conference (GI 2012), and the 9th Canadian Conference on Computer and Robot Vision (CRV 2012). TheProgramCommitteereceived80submissionsfromacrossCanadaand18 othercountries.Eachsubmissionwasreviewedbyfour membersofthe Program Committee, and for the final conference programand for inclusion in these pro- ceedings,23regularpaperswere accepted,for anacceptancerate of29%.These regularpapersareallocated12pageseachinthe presentproceedings.Addition- ally, 18 short papers, with allocation of six pages each, were accepted. Finally, four papers from the Graduate Student Symposium appear in the proceedings, each of which was allocated four pages. WewouldliketothankallProgramCommitteemembersandexternalreview- ersfortheirwonderfuljobatprovidinghigh-qualityreviewsinatimelymanner. Wethankalltheauthorsofsubmittedpapersandtheauthorsofselectedpapers for their collaboration in preparation of these proceedings. The conference program featured two keynote presentations: one by Eduard Hovy,Directorofthe HumanLanguageTechnologyGroupandDeputy Director of the Intelligent Systems Division of the Information Sciences Institute of the University of Southern California, and one by Samy Bengio, Research Scientist in Machine Learning at Google Inc. Many thanks to Ebrahim Bagheri and Jocelyne Faddoul for organizing the Graduate Student Symposium, and chairing its Program Committee. The con- ference benefited from the practical perspective brought by the participants in the Industry Track session. Many thanks to Atefeh Farzindar for organizing it. TheconferencewassponsoredbytheCanadianArtificialIntelligenceAssoci- ation(CAIAC),andwewouldliketotakethisopportunitytothanktheCAIAC ExecutiveCommitteefortheirconstantsupport.Aspecialthankyounotetothe development team of the EasyChair Conference System, which made our work so much easier. Finally, we would like to express our gratitude to John Barron and Cory Butz, the AI/GI/CRV General Co-chairs, Chrysanne DiMarco, Hon- ourary Chair, Michael Jenkin, the AI/GI/CRV Local Arrangements Chair, and Yves Lesprance,the AI LocalOrganizingChair,formaking AI/GI/CRV2012a success. May 2012 Leila Kosseim Diana Inkpen Organization AI 2012 was sponsored by CAIAC, the Canadian Artificial Intelligence Association, and held in conjunction with the 38th Graphics Interface Confer- ence(GI2012),andthe9thCanadianConferenceonComputerandRobotVision (CRV 2012). Executive Committee AI/GI/CRV General Chairs John Barron University of Western Ontario Cory Butz University of Regina AI/GI/CRV Honorary Chair Chrysanne DiMarco University of Waterloo AI/GI/CRV Local Arrangements Chair Michael Jenkin York University AI Program Co-chairs Leila Kosseim Concordia University Diana Inkpen University of Ottawa AI Local Arrangements Chair Yves Lesp´erance York University AI Graduate Student Symposium Co-chairs Ebrahim Bagheri Athabasca University Jocelyne Faddoul Dalhousie University AI Industry Chair Atefeh Farzindar NLP Technologies Inc. and Universit´e de Montr´eal VIII Organization Sponsoring Institutions Canadian Artificial Intelligence Association (CAIAC) http://caiac.ca Department of Computer Science and Software Engineering, Faculty of Engineering and Computer Science, Concordia University http://www.cse.concordia.ca Palomino System Innovations Inc. http://www.palominosys.com NLP Technologies http://www.nlptechnologies.ca Program Committee Esma Aimeur Universit´e de Montr´eal Ahmed Ali Abdala Esmin Universidade Federal de Lavras Dirk Arnold Dalhousie University Ebrahim Bagheri Athabasca University Scott Buffett National Research Council Cory Butz University of Regina Maria Fernanda Caropreso University of Ottawa Eric Charton E´cole Polytechnique de Montr´eal Colin Cherry National Research Council David Chiu University of Guelph Lyne Da Sylva Universit´e de Montr´eal Chrysanne Dimarco University of Waterloo Larbi Esmahi Athabasca University Jocelyne Faddoul Dalhousie University Atefeh Farzindar NLP Technologies Inc. Michel Gagnon E´cole Polytechnique de Montr´eal Yong Gao University of British Columbia, Okanagan Dragan Gasevic Athabasca University Cyril Goutte National Research Council Diana Inkpen University of Ottawa Ilya Ioshikhes University of Ottawa Aminul Islam Dalhousie University Christel Kemke University of Manitoba Vlado Keselj Dalhousie University Fazel Keshtkar University of Memphis Ziad Kobti University of Windsor Grzegorz Kondrak University of Alberta Leila Kosseim Concordia University Adam Krzyzak Concordia University Philippe Langlais Universit´e de Montr´eal Guy Lapalme Universit´e de Montr´eal Organization IX Fuhua Lin Athabasca University Alejandro Lopez-Ortiz University of Waterloo Alan Mackworth University of British Columbia Yannick Marchand Dalhousie University Joel Martin National Research Council Stan Matwin University of Ottawa Gordon Mccalla University of Saskatchewan Marie-Jean Meurs Concordia University Saif Mohammad National Research Council David Nadeau InfoGlutton Jian-Yun Nie Universit´e de Montr´eal Roger Nkambou Universit´e du Qu´ebec `a Montr´eal(UQAM) Doina Precup McGill University Robert Reynolds Wayne State University Frank Rudzicz University of Toronto Mahdi Shafiei Acadia University Mohak Shah Accenture Technology Labs Weiming Shen National Research Council Marina Sokolova University of Ottawa and CHEO Research Institute Bruce Spencer National Research Council Stan Szpakowicz University of Ottawa Choh Man Teng Florida Institute for Human and Machine Cognition Thomas Tran University of Ottawa Peter Van Beek University of Waterloo Chun Wang Concordia University Harris Wang Athabasca University Xin Wang University of Calgary Dunwei Wen Athabasca University Ren´e Witte Concordia University Yang Xiang University of Guelph Jia-Huai You University of Alberta Xiaokun Zhang Athabasca University Nur Zincir-Heywood Dalhousie University Additional Reviewers Asadi, Mohsen Moldovan, Cristian Drummond, Chris Pan, Sinno J. Fatima, Shaheen Peckham, Terry Guo, Jinsong Raahemi, Bijan Jankowska,Magdalena Razavi, Amir H. Klement, William Salameh, Mohammad Liu, Guohua Su, Jiang Mohabbati, Bardia Zhao, Lingzhong Invited Talks Large Scale Semantic Extraction through Embeddings: From Images to Music Samy Bengio Google [email protected] Image annotation datasets are becoming largerand larger,with tens of millions of images and tens of thousands of possible annotations. In the first part of the talk, I’ll introduce WSABIE, a strongly performing method that scalesto suchdatasetsby simultaneouslylearning to optimize pre- cision at k of the ranked list of annotations for a given image and learning a low-dimensional joint embedding space for both images and annotations. Our methodbothoutperformsseveralbaselinemethodsand,incomparisontothem, isfasterandconsumeslessmemory.Wealsodemonstratehowourmethodlearns an interpretable model, where annotations with alternate spellings or even lan- guages are close in the embedding space. Hence, even when our model does not predict the exact annotation given by a human labeler, it often predicts similar annotations. Inthesecondpartofthetalk,I’llshowhowthesameapproach,WSABIE,can be extended to the multi-task case, where one learns simultaneously to embed in the same space variousmusic relatedinformationsuchas artistnames,music genres, and audio tracks in order to optimize different but related costs. A New Semantics: Merging Propositional and Distributional Information Eduard Hovy Information Sciences Institute,University of Southern California [email protected] Despite hundredsofyearsofstudy onsemantics,theoriesandrepresentationsof semantic content—the actual meaning of the symbols used in semantic propositions—remainimpoverished.The traditionalextensionaland intensional models of semantics are difficult to actually flesh out in practice, and no large- scale models of this kind exist. Recently, researchers in Natural Language Pro- cessing (NLP) have increasingly treated topic signature word distributions (also called ‘context vectors’, ‘topic models’, ‘language models’, etc.) as a de facto placeholder for semantics at various levels of granularity. This talk argues for a new kind of semantics that combines traditional symbolic logic-based proposition-stylesemantics (of the kind used in older NLP) with (computation- based) statistical word distribution information (what is being called Distribu- tional Semantics in modern NLP). The core resource is a single lexico-semantic ‘lexicon’ that can be used for a variety of tasks. I show how to define such a lexicon, how to build and format it, and how to use it for various tasks. Com- bining the two views of semantics opens many fascinating questions that beg study, including the operation of logical operators such as negation and modal- ities over word(sense) distributions, the nature of ontological facets required to define concepts, and the action of compositionality over statistical concepts.
The list of books you might like
